Mauro Gaio

Movement descriptions do not only indicate where people go, they also reveal how routes are understood, followed, and explained through actions, landmarks, directions, and distances. While many computational approaches to mobility rely on GPS traces or named place recognition, textual itineraries contain richer spatial cues that are often expressed through ordinary landmarks such as paths, bridges, churches, or intersections. This work studies French hiking descriptions as a case study for extracting structured spatial information from human-written route instructions. We build an annotated corpus of route descriptions and develop neural methods to identify the linguistic elements that support movement understanding, including named places, nominal landmarks, movement actions, spatial relations, and measures. Particular attention is given to nominal spatial entities, which are frequent in route instructions but difficult to match directly to geographic databases. The long-term objective is to support an explainable route reconstruction system able to connect textual movement descriptions with geographic evidence while preserving the way humans naturally describe routes.

This paper presents an approach to GWS (GeospatialWeb Service) discovery through the semantic annotation of WPS (Web Processing Service) service descriptions. The rationale behind this work is that search engines that use appropriate semantic-based similarity measures in the matching process are more accurate in terms of precision and recall than those based on syntactic matching alone. The lack of semantics in the description of services using a standard such as WPS prevents the use of such a matching process and is considered a limitation of GWS discovery. The GWS discovery approach presented is based on the consideration of semantics in the service description method and in the matching process. The description of services is based on a semantic lightweight meta-model instantiated in the WPS 2.0 standard, extending the description of the service through metadata tags. The matching process is performed in three steps (functionality matching step, I/O (Input/Output) matching step and non-functional matching step). Its core is a semantic similarity measure that combines logical and non-logical matching methods. Finally, the paper presents the results of an experiment applying the proposed discovery approach on a GWS corpus, showing promising results and the added value of the three-step matching process.

In this paper, we propose and discuss a methodology to map the spatial fingerprints of novels and authors based on all of the named urban roads (i.e., odonyms) extracted from novels. We present several ways to explore Parisian space and fictional landscapes by interactively and simultaneously browsing geographical space and literary text. Our project involves building a platform capable of retrieving, mapping and analyzing the occurrences of named urban roads in novels in which the action occurs wholly or partly in Paris. This platform will be used in several areas, such as cultural tourism, urban research, and literary analysis. The paper focuses on extracting named urban roads and mapping the results for a sample of 31 novels published between 1800 and 1914. Two approaches to the annotation of odonyms are compared. First, we describe a proof of concept using queries made via the TXM textual analysis platform. Then, we describe an automatic process using a natural language processing (NLP) method. Additionally, we mention how the geosemantic information annotated from the text (e.g., a structure combining verbs, spatial relations, named entities, adjectives and adverbs) can be used to automatically characterize the semantic content associated with named urban roads.

L’annotation sémantique d’information spatiale a pour objectif de repérer des mots ou des syntagmes décrivant des références géographiques (noms de lieux) ainsi que diverses expressions spatiales associées. L’une des plus importantes difficultés pour concevoir un système automatique d’annotation d’un tel type d’information est due aux ambiguïtés liées aux entités spatiales. Une approche modulaire basée sur des services Web a été choisie. La méthodologie proposée repose sur la combinaison d’une étape de prétraitement (analyse morphosyntaxique), d’une cascade de transducteurs, et d’une étape de classification utilisant des ressources du Web des données. Un avantage de cette approche est la possibilité d’obtenir des traitements partiels ou encore de mettre en concurrence certains modules réalisant la même tâche. The semantic annotation of spatial information aims to identify words or phrases describing geographical references (place names) as well as various associated spatial expressions. One of the major difficulties in designing an automatic annotation system for such information is due to ambiguities related to spatial entities. A modular approach based on Web services was chosen. The proposed methodology is based on the combination of a pre-processing step using external morphosyntactic analysers, a cascade of transducers, and a classification step using linked data. An advantage of this approach is the possibility to obtain partial processing or to evaluate competing methods doing the same task.

GéoTopia is an online platform designed to promote the publication, distribution, exchange, discussion, and analysis of archive data. As an original feature, it incorporates functions for the spatial and temporal management of relatively informal documents of complex structure, such as images, drawings, and texts. A map interface allows the archiving, consultation, and handling of georeferenced documents by means of a zoom bar and a timeline. Display by map juxtaposition facilitates the comparison of two documents. Two Web services with adjustable parameters, GéoStream and TempoStream, have been developed to aid users with the spatial and temporal indexing of the texts that they integrate into the GéoTopia platform. The result of the spatial indexing is displayed on the base map in the form of thumbnail images showing the spatial references extracted from the text, which the user can validate, either wholly or in part.

This paper presents a work package realized for the G\'eOnto project. A new method is proposed for an enrichment of a first geographical ontology developed beforehand. This method relies on text analysis by lexico-syntactic patterns. From the retrieve of n-ary relations the method automatically detect those involved in a spatial and/or temporal relation in a context of a description of journeys.

Automatic construction of ontologies from text is generally based on retrieving text content. For a much more rich ontology we extend these approaches by taking into account the document structure and some external resources (like thesaurus of indexing terms of near domain). In this paper we describe how these external resources are at first analyzed and then exploited. This method has been applied on a geographical domain and the benefit has been evaluated.